Opinion: DeFi is for Bitcoin TooBy Kyle Torpey
These days, Ethereum evangelists’ favorite trend is Decentralized Finance (DeFi). The basic idea behind DeFi is to replace pretty much every traditional financial service with some kind of decentralized network or application.
Decentralized exchanges like Uniswap, for example, allow users to trade various tokens issued on top of the Ethereum network without the need for a trusted third party. Another example of DeFi is Maker DAO, which allows users to create a USD stablecoin backed by their ETH holdings — also without placing trust in a third party. As one of the many buzzwords in the cryptocurrency space, the DeFi tag also gets attached to projects that aren’t all that decentralized.
But while DeFi has mostly been promoted as an Ethereum phenomenon, the reality is that Bitcoin can do DeFi too.
Bitcoin as a DeFi Platform
First of all, Bitcoin itself is a form of DeFi. Bitcoin enables, in a decentralized fashion, the traditional financial service of secure value storage and transfers. The Bitcoin network was likely the first real world DeFi app to ever exist, and developers are now also building new DeFi apps on top of the Bitcoin network.
The best example of a DeFi app built on top of Bitcoin may be Abra. Abra’s long term goal is to effectively become a global, non-custodial bank that can be used in a permissionless manner. It’s built on a system of Bitcoin smart contracts, namely multisig addresses. You can find an in-depth explanation of how Abra works here.
Due to Abra’s permissionless and decentralized attributes, users are able to gain access to various types of investments, such as Apple stock or ETFs, for the first time. In a recent survey sent out to their users, Abra found that 43% of their users said accessibility to financial markets, especially U.S. equities, was the main problem the app solved for them. Notably, another 35% of respondents said affordability was the key feature of Abra for them because it allows users to purchase fractional shares of stocks.
Bitcoin’s Limitations with DeFi
It should be noted that building DeFi applications on top of Bitcoin does come with some drawbacks when compared to building on Ethereum. Bitcoin’s scripting language was specifically limited in the early days for security and stability reasons, so developers are unable to be as expressive as they can be with something like Ethereum’s Solidity language.
In other words, there are things developers can do on Ethereum that they can’t do on Bitcoin. Developers on Ethereum are theoretically able to make their DeFi apps require even less trust and enable more decentralization than apps built on Bitcoin.
To illustrate this point, let’s go back to Abra and Maker DAO. Abra and Maker DAO provide a similar service to their users, which is the ability to hedge Bitcoin or Ether with a more stable currency, say the U.S. dollar. But these two DeFi apps achieve this end goal by very different means.
In Abra, a simple 2-of-3 multisig address is used to effectively create a bet between the user and Abra on the price of the U.S. dollar relative to the Bitcoin held in the user’s account. The user goes long USD and Abra goes long bitcoin. A third party (the oracle) decides how much Bitcoin should be sent to Abra and how much Bitcoin should be sent to the user (based on the price movements during the time of the bet) whenever the user decides to withdraw U.S. dollars from their Abra account or simply convert back to Bitcoin. The end result is that the user’s Bitcoin is essentially pegged to the U.S. dollar via the conditions by which that Bitcoin can be sent elsewhere by the user.
Maker DAO is much more complex. For one, a group of many geographically diverse oracles are used instead of one (although more oracles could also be added to Abra in the way Bitcoin exists today). Additionally, the DAI token is generated when a user puts their ETH into the MakerDAO app, which allows for much easier integration with other DeFi apps, especially those built on the Ethereum network. For example, Maker DAO users are able to earn a return on their DAI by lending it out on Compound. There is also the MKR token itself, which is effectively a governance token for the Maker DAO system.
There are many other differences between how Abra and Maker Dao work. These are just some examples.
Other examples of Ethereum’s advantages over Bitcoin in DeFi development include more complex wallets, where things like covenants could be implemented today, and the Lightning Network, where better security guarantees can be offered without the need for new opcodes at the base protocol level (as is the case in Bitcoin).
Building on Bitcoin Has Benefits Too
And yet, building on Bitcoin has benefits too. The network effects around Bitcoin means potentially more users of a DeFi app at launch, and Bitcoin is generally viewed as the most secure and trusted cryptocurrency network in existence due to how long it has been around and the conservative nature of the development process.
In addition to the level of respect the Bitcoin network receives from developers and users, the Bitcoin asset may well be the most trusted form of money in the cryptocurrency space. Although still volatile, Bitcoin has proven itself over many bear markets and has become less volatile over time.
While Bitcoin’s scripting language somewhat limits what developers can build on top of the base network, the idea that Bitcoin doesn’t have smart contracts is incorrect. The Bitcoin community simply takes a different approach to this aspect of blockchain technology. Casa CTO Jameson Lopp explained this difference in smart contract philosophy during an episode of Epicenter earlier this year:
“A lot of the ‘more conservative’ Bitcoin developers don’t like having smart contracts that have to get executed by everyone on the network. They rather want to perform the same type of logic but where the actual execution happens privately, and then you’re just providing a proof of the execution that the rest of the world can verify.”
Put differently, the idea is that the Bitcoin blockchain will act as a trust anchor or court for smart contracts that is only called upon when there is a dispute between users on the network. By only calling upon the blockchain when absolutely necessary, Bitcoin users should be able to gain a higher level of privacy and scalability with their smart contracts because not as much information will be stored in a public database for all time.
Ethereum, on the other hand, has already received criticism for the size of its blockchain due to how much bandwidth will be needed to operate a full node in the future, in addition to the increasing difficulty associated with operating a full node.
Ethereum-based DeFi apps are also sometimes unnecessarily complex. For example, many of the DeFi apps being built on top of Ethereum come with an additional token baked into the protocol that creates the need for users to switch between multiple tokens instead of simply using the native ETH asset. There are, of course, exceptions. For example, Uniswap has basically built what the Bancor ICO was building without the need for an additional token.
Another issue with Ethereum is that it’s still extremely difficult to write secure smart contracts. The most obvious example is that Parity, which was founded by Solidity inventor Gavin Wood, had a buggy multisig contract in its Ethereum client, which lead to more than $150 million worth of losses for various ICO projects and other entities that were using the software to secure their funds. Developers are free to create more expressive smart contracts on Ethereum, but that also comes with additional security risks.
As then Bitcoin Core lead maintainer Gavin Andresen wrote before the launch of Ethereum back in 2014, many of the interesting things that can be done with Ethereum can also be built on top of Bitcoin’s multisig functionality. This is mainly due to the oracle problem in which there usually needs to be some sort of person in the real world making a decision on a smart contract or submitting some data to the blockchain for use in a smart contract.
Smart contracts aren’t as smart or automated as advertised. For example, it’s unclear if Maker DAO is sufficiently decentralized, as the system still relies on trusted price data feeds to function properly. In reality, there would not be much difference in the level of censorship-resistance and decentralization between Abra and Maker DAO if Abra were to add more oracles to its system.
The caveat here is, as mentioned previously, Ethereum enables developers to hypothetically implement lower trust requirements and greater decentralization through a more expressive scripting language. But at this point in time, it’s unclear how useful these additional features will be for DeFi apps. So it’s possible that Bitcoin can offer a sufficient level of decentralization.
It’s also important to point out that Bitcoin is adding the ability to implement things like covenants and improvements for the Lightning Network as well. Additionally, many of these advanced smart-contracting features have yet to be deployed on Ethereum — let alone in a provably secure manner.
The Taproot proposal is the next big step forward for Bitcoin smart contracts, as it offers massive improvements in the areas of privacy, scalability, and flexibility. Having said that, it’s unclear when it will find its way into the protocol. According to comments made by longtime Bitcoin developer Matt Corallo at the recent Bitcoin 2019 conference, Taproot could still be as much as another two years away from activation on the Bitcoin network.
Another key proposal to watch in terms of DeFi on Bitcoin is Drivechain. Drivechain could open the door to the same sort of open innovation that takes place on Ethereum today, albeit in an opt-in manner with different security tradeoffs. RSK is a Bitcoin sidechain meant to emulate much of what Ethereum does that already exists in the wild. This Ethereum-esque sidechain uses a model that combines the Drivechain concept with a Liquid-like federation model.
Drivechain creator Paul Sztorc is also eager to launch his decentralized prediction markets sidechain, known as Bitcoin Hivemind, as soon as possible. Sztorc is the intellectual godfather of Augur and some other prediction market projects on Ethereum, which are also considered part of the Ethereum DeFi ecosystem.
If the Drivechain experiment is successful, it would be somewhat trivial to launch a Bitcoin version of any sort of interesting Ethereum DeFi app as its own separate sidechain (or on top of the RSK sidechain). Although, as mentioned previously, a large number of interesting DeFi applications can already be built on top of Bitcoin as it exists today.
Further down the road, something like the Blockstream-developed Simplicity language could make its way into Bitcoin in some way, which would allow for the development of more sophisticated smart contracts.
Much of this remains theoretical, however. History has shown that it can be quite difficult to get additional features added to Bitcoin, due to the conservative nature of the community. Protecting Bitcoin as “digital gold” still takes priority over everything else.
That said, it’s already possible to build rather powerful DeFi applications on top of Bitcoin as it exists today. Furthermore, while Ethereum allows developers more freedom to experiment with DeFi today, it’s unclear how long that advantage will continue to exist. Bitcoin already serious advantages in the areas of network effects, trust, longevity, and its perception as real money.
So if you want to track DeFi developments, it’s important to look at Bitcoin as well.